Patient-derived dependency maps

Our mission is tooutsmart cancer

We build gene dependency maps from real patient tumors, not cell lines, because the best model of human cancer is the patient. Real tumor biology delivers real targets: synergistic target pairs for bispecific antibody and ADC programs.

Book a scoping call →How the platform works

Motivated to make better drugs·Powered by tumor data·Delivered by machine learning

SCROLL

In numbers

Gene dependency maps from 13,607 patient-derived tumors

Synergistic target pairs for bispecific antibody and ADC programs.
Probability of success before Phase II.

patient tumors in a purpose-built database

tissue types

classes of mutations

gene-altering mutations

RNA-seq and single-cell RNA datasets

accuracy predicting clinical outcomes, retrospective validation

What we do

From patient tumors to drug targets and go/no-go decisions

Every tumor is a natural experiment in which some genes could not be lost. We measure that across thousands of patient tumors, per tissue and per genetic background, and turn it into answers a drug program can act on.

Patient tumor data

Genomic, transcriptomic and clinical data from real patients, not cell lines, organised by tissue and mutation class.

Dependency maps

Machine learning reads which genes, and which gene pairs, a tumor cannot afford to lose. That is the dependency signal.

Answers you can act on

Ranked targets and synergistic pairs for your indication, or a probability-of-success score for an asset you are weighing.

Three ways to work with us ↓

Explore

Three ways to work with us, and our own pipeline

Target identificationDrug targets validated by patient tumorsNovel targets and synergistic target pairs that cell-line screens cannot see, in any indication.Explore →

Asset evaluationProbability of success before Phase IIA translational-risk score for any oncology program; 90% accuracy in retrospective validation.Explore →

Before you investGo/no-go evidence from patient dataEleven investment-lessons reports and due-diligence assessments for investors.Explore →

PipelineOur own bispecific programsThree programs for genetically defined populations. One licensed to Kohde Pharma.Explore →

Team

Built by people who have taken drugs into the clinic

Oncology drug development and computational cancer biology under one roof, with a team and a Scientific Advisory Board drawn from the University of Oxford, Harvard Medical School, Pfizer, Evotec and Omega Funds.

Meet the team →

Partnered and backed by

Contact

Request a target assessment

Tell us the indication, target or asset. We reply within two business days with scope, timeline and a sample output.

Gordion Bioscience Inc.
245 Main St, Cambridge, MA 02142, US
contact@gordion.bio